numexpr

Page 1 of 2

ID Status Summary
129 Done NumPy version is lower than needed: 1.10.0.dev0+0752872 < 1.6 Type-Defect Priority-Medium
128 New Installing numexpr on to my Mac 10.9.4 Maverick Type-Defect Priority-Medium
127 New Numexpr doesn't seem to be installed with MKL through pip Type-Defect Priority-Medium
126 New arr.dtype.type hash differs from numpy hash for SOME dtypes Type-Defect Priority-Medium
125 Accepted IMPORTANT: *** Do not file tickets here anymore! *** Type-Other Priority-Critical
124 New Disable tests that use multithreading on sparc Type-Defect Priority-Medium
123 New numexpr unstable due to use of platform.machine() Type-Defect Priority-Medium
122 Invalid build on mac 10.9, python 3.3 failed Type-Defect Priority-Medium
121 New Numexpr query with empty string does not work (discovered using pytables) Type-Defect Priority-Medium
120 New add min and max functions Type-Defect Priority-Medium
119 Done User's guide seems to have disappeared - can it be restored Type-Defect Priority-Medium
118 New _context threadlocal incompatible with eventlet green threads Type-Defect Priority-Medium
117 New Adding conj function Type-Defect Priority-Medium
116 Fixed Build failure on s390 architecture Type-Defect Priority-Medium
115 Fixed NumExpr 2.2.1 should be compatible with PyTables <= 2.4 Type-Defect Priority-Medium
114 Fixed Fix test failures with numpy 1.8 Type-Defect Priority-Medium
113 New Add support for Yeppp! Type-Defect Priority-Medium
112 Invalid numexpr.evaluate result type != numpy type Type-Defect Priority-Medium
111 Fixed return TestResult in numexpr.test() Type-Defect Priority-Medium
110 Fixed No way to limit the number of threads before importing / please honour OMP_NUM_THREADS Type-Defect Priority-Medium
109 Fixed Missing license files Type-Defect Priority-Medium
108 Invalid Error in FindNumPy.cmake Type-Defect Priority-Medium
107 Fixed Modulus with 0 and integer array causes floating point exception Type-Defect Priority-Medium
106 New int8 or int16 support Type-Enhancement Priority-Medium
105 New add an example of using strings (literals) on Python3 Type-Enhancement Priority-Medium
104 Fixed Patch for /FindNumPy.cmake Type-Patch
103 Fixed result in-place enhancement Type-Enhancement Priority-Medium
102 Fixed better library suffix finding in FindPythonLibsNew Type-Defect Priority-Medium
101 Invalid Can't install numexpr, creating dist error: could not create 'dist': Permission denied Type-Defect Priority-Medium
100 New numexpr-2.1 does not build with Python 3.2 (builds fine with Python 2.7) Type-Defect Priority-Medium
99 New Solve for MKL_Get_Version_String while build numexpr-2.1 with Intel MKL VML Type-Defect Priority-Medium
98 Invalid python segfault from doubly defined procedure + numexpr Type-Defect Priority-Medium
97 Fixed only one core used when operating on arrays returned by np.ogrid Type-Defect Priority-Medium
96 New Algebra for boolean values doesn't work on arrays (interpret as 0 / 1 integer) Type-Defect Priority-Medium
95 Fixed Solve for MKLGetVersionString while using Intel MKL VML support in numexpr-2.0.1 Type-Defect Priority-Medium
94 Fixed can't build with vml on linux Type-Defect Priority-Medium
93 New Input/output to same array should work Type-Defect Priority-Medium
92 New where() does not work with booleans and strings due to check in numexpr.expressions.commonKind Type-Defect Priority-Medium
91 Fixed NumExpr and Python 3 Type-Defect Priority-Medium
90 New ImportError: cannot import name _get_vml_version Type-Defect Priority-Medium
89 New Add np.trunc, np.ceil equivalents Type-Defect Priority-Medium
88 New Make numexpr installable pip Type-Defect Priority-Medium
87 New Add np.sign function Type-Defect Priority-Medium
86 New Add minimum, maximum functions Type-Defect Priority-Medium
85 New multidimensional output Type-Defect Priority-Medium
84 New give option to use optimized object Type-Defect Priority-Medium
83 New Can evalute expersions with record arrays Type-Defect Priority-Medium
82 Invalid AttributeError: 'OpNode' object has no attribute 'sum' Type-Defect Priority-Medium
81 New numexpr.evaluate('(a*1j)+0') Type-Defect Priority-Medium
80 New numexpr is not thread-safe Type-Defect Priority-Medium
79 New sum and prod incorrect results Type-Defect Priority-Medium
78 New support accumulate Type-Enhancement Priority-Medium
77 Accepted failing unittests / crashes on GNU/Linux Debian sparc arch (in threaded mode) Type-Defect Priority-Medium
76 Accepted -sum(a) does not work Type-Defect Priority-Medium
75 Fixed Test failure on Hurd and KFreeBSD Type-Defect Priority-Medium
74 New Full support of Intel VML functions Type-Defect Priority-Medium
73 New calculation of sum is slow and uses only one core Type-Defect Priority-Medium
72 New Test failure of 1.4.2 on 32bit platforms Type-Defect Priority-Medium
71 Accepted float constant changes to integer in a**-1.0 Type-Defect Priority-Medium
70 Invalid incorrect usage of inetger division Type-Defect Priority-Medium
69 Fixed Reduction failure Type-Defect Priority-Medium
68 Accepted ones_like, copy and fmod are undocumented Type-Defect Priority-Medium
67 Accepted Numexpr 2.0 only accepts 31 different variables as input Type-Defect Priority-Medium
66 Fixed extension to necompiler.evaluate breaks Python 2.4 + 2.5 compatibility Type-Defect Priority-Medium
65 Fixed MAX_THREADS not large enough Type-Defect Priority-Medium
64 Verified numexpr 2.0rc1 does not accept empty arrays as input Type-Defect Priority-Critical
63 Fixed Wrong result produced when input array has a negative stride Type-Defect Priority-Medium
62 Fixed Licence boilerplates missing Type-Defect Priority-Medium
61 New where function with implicit booleans Type-Defect Priority-Medium
60 WontFix Support Indexing Type-Enhancement Priority-Medium
59 Verified floats without decimals are evaluted as integers Type-Defect Priority-Low
58 Verified Explicit type conversion functions Type-Enhancement Priority-Medium
57 New Add conjugate() function Type-Enhancement Priority-Low
56 Fixed Mechanism to store expressions in already existing ndarrays Type-Enhancement Priority-Medium
55 Fixed Trivial typographic error Type-Defect Priority-Medium
54 Accepted Unable to build on OSX 10.7 when linking with MKL Type-Defect Priority-Medium
53 Fixed Patch for /trunk/numexpr/cpuinfo.py Type-Patch
52 Fixed cpuinfo.getoutput output isn't defined Type-Defect Priority-Medium
51 Verified Is building with MSVC possible? Type-Defect Priority-Medium
50 Fixed trunk fails to build on MingW Type-Defect Priority-Medium
49 Fixed Slow down compared to numpy Type-Defect Priority-Medium
48 Invalid Build fails on OSX 10.6.6 Type-Defect Priority-Medium
47 Fixed numexpr does not parallelize well long column, short row size array Type-Defect Priority-Medium
46 WontFix support for class instant variables. (self.abc) Type-Enhancement Priority-Medium
45 WontFix Numexpr fails most tests with python 2.4 Type-Defect Priority-Medium
44 Fixed possible build problem with gcc 4.5.2? Type-Defect Priority-Medium
43 Verified Optimization of "compilation phase" through a better hash Type-Defect Priority-Medium
42 Verified Potential missing Py_END_ALLOW_THREADS Type-Defect Priority-Medium
41 Verified sum big array causes segmentation fault Type-Defect Priority-Medium
40 Invalid numexpr crashes python on win32 (numexpr-1.4.1.win32-py2.6.exe) Type-Defect Priority-Medium
39 Verified The number of threads should be set only for Numexpr native pthreads, not for VML Type-Enhancement Priority-Medium
38 Invalid [PATCH] Loop optimization to gain ~15% performance Type-Defect Priority-Medium
37 Verified Raising array containing 0s to a negative power causes core dump in numexpr 1.4 Type-Defect Priority-Medium
36 Verified Multi-threading code should be disabled for small arrays Type-Defect Priority-Medium
35 Fixed Numexpr should release the GIL during computations Type-Enhancement Priority-High
34 Fixed No AMD64 builds for numexpr 1.4 and Python 2.7 Type-Defect Priority-Medium
33 Verified Numexpr does not work correctly when run in sub-processes Type-Defect Priority-Medium OpSys-All
32 WontFix symbol lookup error: /usr/pkg/fortran90/intel/mkl/10.2.5.035/lib/em64t/libmkl_vml_mc2.so: undefined symbol: mkl_serv_mkl_malloc Type-Defect Priority-Medium
31 Fixed numexpr 1.4 does not build (as is) on Mingw Type-Defect Priority-Medium
30 Fixed Broadcasting does not work for parallel code correctly Type-Defect Priority-Medium OpSys-All